Centre for Multi-Dimensional Data Visualisation (MuViSU)


EMS Research day 2025: 6 November 2025


muvisu@sun.ac.za

Who we are

  • MuViSU is a research centre located within the Department of Statistics and Actuarial Science at Stellenbosch University
  • Established in 2021.
  • Core activities:
    • To extend multi-dimensional visualisation methodology and related techniques, such as biplots, through theoretical developments;
    • The application of newly derived techniques to data sets originating from various fields;
    • The development, maintenance, and improvement of an extensive collection of R functions for constructing all necessary graphical displays and performing various multi-dimensional visualisation based techniques.

Members

  • Institutional member: School for Data Science and Computational Thinking (SU)
  • Affiliations of 25 individual members:
Stellenbosch University: 7 Erasmus University: 3 Universitat Pompeu Fabra: 1
North West University: 2 Universiteit Leiden: 2 Univerisity Compania “Luigi Vanvielli”: 1
University of the Witwatersrand: 1 Democritus University: 1 University of Naples Federico II: 1
University of the Free State: 1 Universite De Moncton: 1 University of Wollongong: 1
Sasol: 1 TU Wien: 1 Independent risk analyst: 1
  • 1 Postdoctoral Research Fellow
  • Students:
    • 4 PhD students
    • 1 Masters student
    • Graduated: 5 Masters

Management Committee

Director:
Sugnet Lubbe (SU)

Deputy-Director:
Johané Nienkemper-Swanepoel (SU)

Secretary:
Raeesa Ganey (Wits)

Financial Officer:
Roelof Coetzer (NWU)

Software

biplotEZ - pronounced biplot easy

  • EZ-to-Use Biplots.
  • Provides a unified approach to constructing various biplots.

Lubbe S, le Roux N, Nienkemper-Swanepoel J, Ganey R, Buys R, Adams Z, Manefeldt P (2025). biplotEZ: EZ-to-Use Biplots. R package version 2.3, https://github.com/MuViSU/biplotEZ.

moveEZ - pronounced move easy

  • Animated Biplots.
  • Calls biplotEZ to create biplot objects.

Ganey R & Nienkemper-Swanepoel J (2025). moveEZ: Animated Biplots. R package version 1.1.1, https://CRAN.R-project.org/package=moveEZ.

wideRhino - pronounced wide Rhino

  • High-Dimensional Methods via Generalised Singular Decomposition.
  • Suitable to visualise data with a large number of variables compared to samples.

Ganey R (2025). wideRhino: High-Dimensional Methods via Generalised Singular Decomposition. R package version 1.0.2, https://CRAN.R-project.org/package=wideRhino.

bipl5 - pronounced bipl vyf

  • Interactive biplot visualisation.
  • Reactive Calibrated Axes.

Buys R & van der Merwe C (2023). bipl5: Construct Reactive Calibrated Axes Biplots. R package version 1.0.2, https://CRAN.R-project.org/package=bipl5.

GPAbin - pronounced G - P - A - bin

  • Mulitple imputation of multivariate categorical data.
  • Unifying Multiple Biplot visualisations into a Single Display.

Nienkemper-Swanepoel J (2025). GPAbin: Unifying Multiple Biplot Visualisations into a Single Display. R package version 1.1.1, https://CRAN.R-project.org/package=GPAbin.

ClusBoot - pronounced clusboot

  • Bootstrap a Clustering Solution to Establish the Stability of the Clusters.

Lubbe S (2024). ClusBoot: Bootstrap a Clustering Solution to Establish the Stability of the Clusters. R package version 1.2.2, https://CRAN.R-project.org/package=ClusBoot.

biplotEZ

library(biplotEZ)
bp <- biplot(Africa_climate, scaled = TRUE) |> 
  PCA(group.aes = Africa_climate$Region) |> 
  samples(opacity = 0.8, 
          col = scales::hue_pal()(10)) |> 
  axes(col="black") |> 
  plot()

moveEZ

bp |> moveplot(time.var = "Year", group.var = "Region", hulls = TRUE, move = TRUE)

bipl5

Biplots for different data types

  • Principal Component Analysis (PCA) biplot: continuous variables
  • Categorical PCA (CatPCA) biplot: numeric and categorical variables
  • Correspondence Analysis (CA) maps: two categorical variables (frequency tables)
  • Multi-Dimensional Scaling (MDS) biplots: more general distance based methods
  • Grouped data:
    • Canonical Variate Analysis (CVA) biplots
      • More variables than samples (wideRhino)
    • Analysis of Distance (AoD) biplots
  • Features:
    • 1D, 2D, 3D biplots
    • Fit measures to understand success of dimension reduction and lower dimensional representation
    • Visual summaries: alpha-bags, densities
    • Multiple Correspondence Analysis (MCA) biplots: multiple categorical variables (GPAbin)
  • moveEZ and bipl5 current versions only include PCA biplots.

Thank you | Enkosi | Dankie